37 votes

Study finds people are consistently and confidently wrong about those with opposing views

28 comments

  1. [2]
    creesch
    Link
    I feel like this is more prevalent or visible in political discourse where people feel the "stakes" are higher. From various sides of the spectrum, you see people make extremely confident...

    I feel like this is more prevalent or visible in political discourse where people feel the "stakes" are higher. From various sides of the spectrum, you see people make extremely confident statements about other sides that are absolutely wrong. Although I feel that the more polarized a subject is, the more we actually start thinking in caricatures of the other side. Without even realizing it, we do this.

    Outside of politics, I feel like this behavior is also visible, but in a more subtle way. You see it often enough in online discourse, where one or both sides more or less responds to what they assume the other thinks. Rather than replying to the contents of what someone wrote.

    32 votes
    1. Raspcoffee
      Link Parent
      Yeah, and I'm well aware that I've often fallen victim to this myself. It's easier to assume something, and generally speaking it is easier to respond than to actively converse. Even more nasty...

      Outside of politics, I feel like this behavior is also visible, but in a more subtle way. You see it often enough in online discourse, where one or both sides more or less responds to what they assume the other thinks. Rather than replying to the contents of what someone wrote.

      Yeah, and I'm well aware that I've often fallen victim to this myself. It's easier to assume something, and generally speaking it is easier to respond than to actively converse. Even more nasty and more subtle, if the conversation steers a different direction that way you can - without noticing - end up in a (sub)topic where you have a greater advantage due to having more knowledge on something. In the original paper they mention something about using less words, and less complex words for mental states of the 'out group'. I wonder if that's correlated.

      Reading the original paper, we also appear to overestimate our understanding of our 'own side', but to a lesser degree. Which is also a valuable insight. On many points we may disagree more than we think on one end of the spectrum, in the same way we misunderstand the other end.

      12 votes
  2. vord
    (edited )
    Link
    I haven't dug in yet, but I suspect this study falls prey to the 'sanitized question' problem. It's that same phenomenon that makes the political compass kinda garbage. I recall (but can't find) a...

    I haven't dug in yet, but I suspect this study falls prey to the 'sanitized question' problem. It's that same phenomenon that makes the political compass kinda garbage. I recall (but can't find) a poll that showed overwhelming support for every policy in Medicare For All when abstracted from the policy name, but as soon as Medicare was mentioned conservative support plummetted.

    I know many right-wing people who will happily declare that immigration is good while simultaneously arguing that illegal immigrants need to be shot on sight for border crossing.

    I wonder how many questions were more like 'Illegal immigrants should be deported even if their kids are citizens.' Bet that one has a lot more predictive power. If it doesn't, then conservatives really need to learn more about what their party pushes for legislatively.

    The real answer is that people don't sort nicely into left-right buckets, but are forced to do so by a shit voting system.

    19 votes
  3. [2]
    entitled-entilde
    Link
    I don’t really understand how this study works. The very fact that you can create two groups suggests beliefs are correlated. Because the participants were clustered by political belief according...

    I don’t really understand how this study works.

    In all, 256 participants were recruited from the U.S. and split evenly between those with left- and right-leaning political views.

    The very fact that you can create two groups suggests beliefs are correlated.

    For each statement, the participant would then be presented with someone else's response to the same statement.

    The participant was then asked to predict the other person's response on a second statement

    Because the participants were clustered by political belief according to the study design, this must be possible. Maybe participants struggled to understand the trade offs of asking for more information (which reduced the payout) or rating their confidence (people struggle with such scales), but they should be able to guess with more than random chance.

    Maybe someone with a better background in psychology could digest the study better though, I’m sure I’m missing some nuance if the measure.

    14 votes
    1. krellor
      Link Parent
      I didn't know if it answers your question but the abstract of the paper is more succinct and clear to me. Paper in nature

      I didn't know if it answers your question but the abstract of the paper is more succinct and clear to me.

      Paper in nature

      To understand these effects, one must examine the cognitive processes underlying how people think about others. Here, we investigate whether people are less prone to theorise about the minds of out-groups, or less able to do so. Participants (Study 1: n = 128; Study 2: n = 128) made inferences about social and political beliefs held by real in-group and out-group members, and could choose to receive further information to improve these inferences. Results show: (1) participants sought equivalent or greater information about out-groups relative to in-groups; but despite this, (2) made significantly less accurate inferences for out-groups; and (3) were significantly less aware of their reduced ability.

      11 votes
  4. [18]
    krellor
    Link
    Unfortunately, this is what people increasingly avoid doing. People are so caught up in in-group/out-group dynamics they lose sight or interest in learning about people, their past, and why they...

    "While there is no quick fix in a real-world setting, if everyone interacted with a more diverse group of people, talked directly to them and got to know them, it's likely we would understand each other better. Conversations with people who hold different beliefs could help challenge our incorrect assumptions about each other."

    Unfortunately, this is what people increasingly avoid doing. People are so caught up in in-group/out-group dynamics they lose sight or interest in learning about people, their past, and why they think what they think.

    12 votes
    1. [15]
      GoodhartMusic
      Link Parent
      I'd argue people don't socialize in general. Community evisceration by digital sales, social media, alarmist media, and siphoning of local funds.

      Because the participants were clustered by political belief according to the study design, this must be possible. Maybe participants struggled to understand the trade offs of asking for more information (which reduced the payout) or rating their confidence (people struggle with such scales), but they should be able to guess with more than random chance.

      I'd argue people don't socialize in general. Community evisceration by digital sales, social media, alarmist media, and siphoning of local funds.

      6 votes
      1. [14]
        krellor
        Link Parent
        Or people socialize online in communities that self-homogenize.

        Or people socialize online in communities that self-homogenize.

        7 votes
        1. [13]
          raze2012
          Link Parent
          Don't all groups by near psychological definition self-homogenizing? That's what makes the nazi bar metaphor so dangerous. You keep a radical group in an area, and the non-radicals will opt out of...

          Don't all groups by near psychological definition self-homogenizing? That's what makes the nazi bar metaphor so dangerous. You keep a radical group in an area, and the non-radicals will opt out of that group. Or are pushed out directly or indirectly.

          So the group either dies out or (more often) gathers more people who either believe in the statement or simply want any sense of belonging or community. The latter is what really makes this go out of control, as people who don't really belive in a cause will fight for it simply because they want to keep up with their "friends". The concept of radicalization is extremely similar to the types of people cults target, because... they kind of are cults, by definition.

          a group requiring unwavering devotion to a set of beliefs and practices which are considered deviant outside the norms of society, which is typically led by a charismatic and self-appointed leader who tightly controls its members.


          In some ways I do agree with GoodHeartMusic that part o this vulnerability comes from the fact that we don't truly "socialize" anymore. When you start to (really) care for no one but yourself, you become vulnerable to anyone who can "relate to you". i.e. anyone who says what you want to hear instead of what you need to hear. I don't really know how that can be resolved in more advanced years; ideally factors like "you're not perfect" and "comproise with friends" should be common social lessons from experience. But that seems to be less and less so.

          4 votes
          1. [11]
            krellor
            (edited )
            Link Parent
            I think it's complicated. Anecdotally, IRL spaces seem less prone to homogenizing. Life is clumpy, and you get local and regional trends, such as rural vs urban. But in my experience, your average...

            I think it's complicated. Anecdotally, IRL spaces seem less prone to homogenizing. Life is clumpy, and you get local and regional trends, such as rural vs urban. But in my experience, your average pub has a broader range of viewpoints across its patrons then your average online community. (Note that I don't participate in many online groups anymore so maybe my info is dated).

            But anecdotally, I can strike up a conversation with someone at a pub, or a conference, on a plane, etc, and be exposed to a much broader set of opinions than online groups, even if I'm in an area or industry with general trends.

            Certain examples, like the Nazi bar one, I think are generally exceptions because they are viewed as pariahs in much of society, so you either radicalize or kick them out.

            Outside of those extremes, you will find liberal, progressive, and conservative views on most most physical gathering places that aren't founded on ideological grounds.

            Again, that is anecdotally, so take it for what it's worth.

            I'll also note that online I've noticed a number of people who will completely tune out a news source or another user if they have even one story or one opinion that they disagree with, even on relatively minor points within a broader area of agreement. I don't remember running into that 20-30 years or so ago, online or offline, so it feels like things have been changing and online tools have made it easy to over filter differences of opinion.

            /Old man yells at cloud.

            5 votes
            1. [2]
              raze2012
              Link Parent
              Well, the nazi example is extreme but also hints at a major difference: moderation. Every hangout place is owned by a businessmen or their trusted hires that know what mood to keep. And the...

              Certain examples, like the Nazi bar one, I think are generally exceptions because they are viewed as pariahs in much of society, so you either radicalize or kick them out.

              Well, the nazi example is extreme but also hints at a major difference: moderation. Every hangout place is owned by a businessmen or their trusted hires that know what mood to keep. And the population of a physical space is naturally limited. There is of course much more friction between facing a bouncer and/or police than re-creating a new account if you get banned.

              Online moderation is less focused on setting a tone and more on appeasing advertisers. And of course the ratio of moderation to malice is much smaller. I empathize a bit with the scale of a large social media, but I also know they could definitely do better. It's just not profitable to do so.

              Outside of those extremes, you will find liberal, progressive, and conservative views on most most physical gathering places that aren't founded on ideological grounds.

              I definitely agree. I'm in a pretty liberal area and it's still not that unusual to run into one or two unique trump hats a day. It probably shifts heavily even from city to city.

              1. krellor
                Link Parent
                Agreed! Thanks for the conversation!

                Agreed! Thanks for the conversation!

                1 vote
            2. [8]
              DefinitelyNotAFae
              Link Parent
              Online homogeneity is also I think common because minority views will get pushed out indirectly and with full denial of it happening. And I'm not talking about political views here even, but PoC,...

              Online homogeneity is also I think common because minority views will get pushed out indirectly and with full denial of it happening. And I'm not talking about political views here even, but PoC, queer, and gender minority voices can be pushed out easily. Sometimes that's the point of an online microspace. But spaces that claim to be all-welcoming spaces will commonly talk over and down to voices they claim to welcome and support.

              This happens IRL but there's more friction there, like you said, and it's easier for people to walk away from unwelcoming online spaces, perpetuating an idea of "this is open to everyone, those sorts of folks just don't want to stay here."

              I find balancing the desire to not be homogenous with the desire not to promote harmful rhetoric one that's too often a tightrope.

              1. [7]
                krellor
                Link Parent
                I agree on the causes of the homogeneity. Speaking purely personally, but I suspect the causes of any harms (as I see them) associated with homogenous spaces stem from whether one has any variety...

                I agree on the causes of the homogeneity. Speaking purely personally, but I suspect the causes of any harms (as I see them) associated with homogenous spaces stem from whether one has any variety in their social diet, so to speak. If someone only socializes and engages in homogenous environments, and worse, only consume news from worldview reinforcing news sources, I think it conditions them to retreat further and further from anything different, or to interpret disagreement with malice (setting aside the obvious hateful rhetoric or hate speech examples, that almost everyone rightly avoids.)

                For me it's more an issue of extent than a binary homogenous vs heterogenous. One person might visit queer or PoC spaces online, while also visiting more general purpose spaces or socializing offline, and keep their social muscles exercised.

                1 vote
                1. [6]
                  DefinitelyNotAFae
                  Link Parent
                  I don't think it's binary either, but I think I worry more about pushing out minority voices, even through passive unwelcoming behavior. Because it harms the minority who lose access to the...

                  I don't think it's binary either, but I think I worry more about pushing out minority voices, even through passive unwelcoming behavior. Because it harms the minority who lose access to the theoretically diverse space which leads to that retreat, but it also actively harms the majority too. The majority voices don't have the microspaces to hear those PoVs. Both are harmed by losing access to the voices they don't hear or the spaces they can't engage in.
                  I know my spaces are richer for hearing Black voices, for example. I have my views challenged and learn more than I do in most heavily white spaces. My Tiktok feed is keyed to hearing folks who will challenge me, often from the left more than the right, but from entire other axes - Indigenous, Black, Asian, etc.

                  I've commented before about how "viewpoint diversity" means having all sorts of diversity, not just hitting a theoretical left and right balance point (which also isn't real).

                  I prefer we as a site not platform racists and I've been really open about that, because I think if there's not the same points being made by non-racists then we probably should be talking about racist they are that they're propagating racist narratives. Because that is also going to make this space less accessible to people explicitly harmed by those narratives (Just like LGBTQ folks don't love only negative news or transphobic comments either and will leave if constantly subjected to them). But I also always try to engage fairly with any commenter or topic, or at least if I absolutely cannot, I'll be very open about my bias or my mood. And I think that transparency is what helps spaces talk about tough things without needing to shut it down.

                  Idk I think I've rambled. Sorry ಠಿ⁠_⁠ಠಿ I think I had a point in there. If it doesn't make sense, please blame the past week I've had

                  1 vote
                  1. [5]
                    krellor
                    Link Parent
                    Eh, no worries, rambling happens and not every conversation needs a "point" beyond a simple exchange of thoughts or ideas. Hope your week calms down! (Note: unless directly referencing your...

                    Eh, no worries, rambling happens and not every conversation needs a "point" beyond a simple exchange of thoughts or ideas. Hope your week calms down!

                    (Note: unless directly referencing your comments, I'm using the term "you" as a generic "someone" not you personally. But it's late and I'm going to bed!)

                    You've mentioned it a bit with your comment about the theoretical left-right balance, and I think that gets to the heart of my personal thoughts. Namely, there is no objectively perfectly balanced space, because life is clumpy and varied and just doesn't work like that. Rather than one perfect space, we should be exposed to many imperfect spaces, and news sources, and people.

                    What I see as a "harm" of homogenous views is the erosion of an individuals ability to thrive in the variety of spaces the world offers, to become radicalized and promote antisocial views, or otherwise withdraw from productive society (not talking workforce productive, cultural engagement is productive too).

                    With regards to individuals thriving, I find it sad that people seem to increasingly withdraw from situations that they might otherwise grow from being exposed to and go on to change the world or achieve big dreams. You don't change the world in 40 hours a week or from a place of comfort. (Not encouraging side hustles; but talking about taking those little plunges)

                    So I'd like to see people at least visit and see all sorts of social compositions, some comfortable, others less so, but in all cases learn to thrive and hold your own, at least for a while. Because unless you go full hermit, life will eventually throw all those situations at you. The difference is whether you thrive or not.

                    Cheers, and you (you you) have a good night!

                    1 vote
                    1. [4]
                      DefinitelyNotAFae
                      Link Parent
                      I think my point is we will usually focus on either a singular axis of "balance" clumpy or otherwise. And I think we won't notice that we're creating an unfriendly space if it isn't ourselves that...

                      I think my point is we will usually focus on either a singular axis of "balance" clumpy or otherwise. And I think we won't notice that we're creating an unfriendly space if it isn't ourselves that are being made uncomfortable.

                      Minorities generally must exist in those less comfortable spaces all the time. Online microspaces create safety for sure, but I think that's why my focus goes more towards that sort of inclusivity, or our larger "neutral" spaces become and remain full of majority viewpoints and thus become antisocial by virtue of perpetuating white/cis/hetero normativity. True radicalization happens, IMO, in those dedicated toxic microspaces for it. But the first steps happen in those macrospaces, where those who are radicalized in toxic microspaces will remain welcome and those minority voices are slowly pushed out. Many subreddits do this, and I've seen complaints from others of Tildes doing it as well, though I think it's gotten better since then, I'd agree with those folks that it isn't always true.

                      Anyway yes my week probably has to get better... Though we started the new one with a very late night ER trip and surprise back pain (unrelated incidents) so who can say.

                      1. [3]
                        krellor
                        Link Parent
                        Right, you don't notice the water when you are a fish. Which is why I don't think micro spaces are bad. Just like in real life where we spend a lot of our time within our comfort zone, that's how...

                        Right, you don't notice the water when you are a fish. Which is why I don't think micro spaces are bad. Just like in real life where we spend a lot of our time within our comfort zone, that's how we spend much of our time online.

                        But if we never leave our comfort zone, we don't grow. And if we let our comfort zone, or maybe in this case tolerance zone, shrink then we regress socially.

                        1 vote
                        1. [2]
                          DefinitelyNotAFae
                          Link Parent
                          Yeah I just think we're coming at the same conclusion from different angles with different lenses! I'm more worried about the spaces, vs the individual's response to them. I'm focusing more on...

                          Yeah I just think we're coming at the same conclusion from different angles with different lenses! I'm more worried about the spaces, vs the individual's response to them. I'm focusing more on community accountability vs individual. And I think both are needed!

                          1. krellor
                            Link Parent
                            Agreed, and thanks for the discussion!

                            Agreed, and thanks for the discussion!

                            1 vote
          2. GoodhartMusic
            Link Parent
            I agree a lot with compromise with friends. When online socialization offers so little friction (beyond being angry someone’s wrong/disagrees — which allows the opportunity to vent your...

            I agree a lot with compromise with friends. When online socialization offers so little friction (beyond being angry someone’s wrong/disagrees — which allows the opportunity to vent your frustration directly to them without consequence) the imposition of compromise may not be as readily accepted.

            4 votes
    2. [2]
      ChingShih
      Link Parent
      I agree. Some of those in-group dynamics work on inertia, though, be it cultural or social, and they help keep up the barriers that prevent discussion and open-mindedness that would promote...

      I agree. Some of those in-group dynamics work on inertia, though, be it cultural or social, and they help keep up the barriers that prevent discussion and open-mindedness that would promote diverse thinking. I am definitely not defending the individual's inability to change their thinking, but we need to also address the historical roadblocks and cultural momentum that's keeping these people on the divisive and defensive course that they're on.

      I feel like in part because of the ease of communication and polarization these days that the politicization and really the weaponization of hate has also spread rapidly to adjacent topics, like what creesch was saying. At the same time, we're exporting much of this cultural polarization and preservation of momentum of some of these negative things and helping this divisiveness to nurture itself in other places where cultural isolation or xenophobia already has deep roots.

      3 votes
      1. krellor
        Link Parent
        I think a lot of it is that there is less in person communication as a proportion of total communication, and online socialization allows for the things you describe like cultural isolation.

        I think a lot of it is that there is less in person communication as a proportion of total communication, and online socialization allows for the things you describe like cultural isolation.

        3 votes
  5. Jordan117
    Link
    I reckon the issue here is that pure ideologues are relatively uncommon and that most people have one or more heterodox views even if they lean one way. Party polarization and zero-sum elections...

    I reckon the issue here is that pure ideologues are relatively uncommon and that most people have one or more heterodox views even if they lean one way. Party polarization and zero-sum elections drive people (especially single-issue voters) to more consistently support one party, which leads to assuming that everybody on the other side is equally monolithic, even though they, like you (probably), only support their team because they're the lesser evil, not because they agree with 100% of the party line.

    11 votes
  6. [2]
    PleasantlyAverage
    Link
    A ~40% accuracy isn't exactly something I would describe as that. The in-group predictions were better at ~50%, but also not "consistently correct".

    even though participants were prepared to seek out [...] information about someone they disagreed with, their predictions were consistently incorrect

    A ~40% accuracy isn't exactly something I would describe as that. The in-group predictions were better at ~50%, but also not "consistently correct".

    8 votes
    1. skybrian
      Link Parent
      Another way to think about it is that they are more confident than they should be, given their error rate.

      Another way to think about it is that they are more confident than they should be, given their error rate.

      2 votes
  7. daywalker
    (edited )
    Link
    I'm going to speculate. I feel like this is the result of a trade-off between different functions. Yeah, when evaluated from a removed place, they are wrong beliefs. However, these beliefs...

    I'm going to speculate. I feel like this is the result of a trade-off between different functions. Yeah, when evaluated from a removed place, they are wrong beliefs. However, these beliefs function to increase in-group cohesion, confidence in the in-group, and it allows one to more easily antagonize and get into conflict with the out-group. With this being said, it can also be said that this can also work against the person, and result in costly conflicts that are avoidable. It can also be speculated that this logic might be becoming more of a problem in the contemporary world, which is very interconnected, interdependent, and complex.

    This is all speculation, but in recent times I've started considering the functional aspects of irrational beliefs and behaviors, and I think considering this aspect has merit. World is not a truth-focused discussion club, and every belief about humans takes place in a social context.

    2 votes
  8. pete_the_paper_boat
    Link
    Pretty good definition of polarization if u ask me

    Pretty good definition of polarization if u ask me